Boron–manganese–carbon nanocomposites synthesized from CO₂ for electrode applications in both supercapacitors and fuel cells

Abstract

This paper describes the derivation of boron–manganese–carbon nanocomposites by CO₂ carbonization using sodium borohydride (NaBH₄) as a reduction agent at 1 bar, followed by impregnation of boron-doped porous carbon (BPC) with a form of manganese oxide (MO). The prepared composites (BPCMO) can be used as an advanced electrochemical energy material, such as active electrocatalysts for oxygen reduction reaction (ORR) and as electrode materials for supercapacitors. Various spectroscopic and microscopic measurements were carried out to investigate the morphology and structure of the BPCMO. Among many types of manganese oxide, it was confirmed that only Mn₃O₄ was embedded in the BPC. Cyclic and linear sweep voltammetry indicated that the BPCMO exhibits a four electron transfer pathway and has electrocatalytic activity comparable to that of commercial Pt/C. Galvanostatic charge/discharge and electrochemical impedance spectroscopic measurements showed that the BPCMO provided remarkable capacitance (150 F g⁻¹ at 1.0 A g⁻¹ and 136 F g⁻¹ at 10.0 A g⁻¹) compared to that of BPC (58 F g⁻¹ at 1.0 A g⁻¹ and 15 F g⁻¹ at 10.0 A g⁻¹), with a highly stable capacitance retention of 93.9% over 3500 charge/discharge cycles. It was found that impregnation of BPC with Mn₃O₄ enhanced electrochemical performance by generation of new active sites, increase in specific surface areas, and reduction of overall resistance.

---

Introduction

The fossil fuels currently used as primary energy sources are becoming scarce and their extensive utilization in modern society has caused global warming due to CO₂ emissions.^1–4 In addition, the ever-growing market for portable electronic devices has created great demand for reliable new power sources. Such recent trends have stimulated interest in research on clean, sustainable energy sources.^5–7 Among many types of renewable energy sources, electrochemical energy devices such as fuel cells, supercapacitors and lithium-ion batteries are promising candidates.⁸ In addition to high energy/power densities, the potential commercialization of these devices entails low costs and high performance levels. These conditions depend on the chemical and physical properties of the materials used for electrodes. Considerable effort has been devoted to satisfying the requirements for such materials by either developing new materials or modifying existing ones.^1,6

In particular, carbon materials provide attractive features such as versatility, good electrical conductivity, chemical stability, large specific surface area, high porosity and low cost.^9–12 Up to now, a myriad of approaches have been reported: nature-inspired materials such as bamboo,¹³ peanut shells,¹⁴ cotton,¹⁵ watermelon,¹⁶ human hair¹⁷ and microalgae¹⁸ have been used as carbon sources for the synthesis of porous carbon. While these attempts have shown reasonable improvements of electrode materials in supercapacitor and fuel cell applications, such source materials are not always easy to come by. Meanwhile, carbon dioxide (CO₂) is a ubiquitous source that surpasses other candidates in terms of accessibility and abundance. Such abundance has led to enormous academic interest in not only reducing CO₂ emissions, but also converting CO₂ to carbon materials using boron-containing hydrides.^19–22 As a result, the CO₂ conversion field has received much public attention in contemporary society.²³

In previous studies, it has been reported that boron-doped porous carbon materials can have a positive effect on electrocatalytic activity.^20,24–30 In particular, the doping of heteroatoms such as phosphorous,³¹ nitrogen³² and sulfur³³ into carbonaceous materials is one strategy to improve electrochemical performance. The effect of heteroatom-doping has been attributed to the differences in electronegativity between carbon and the heteroatoms. An atom that has relatively higher electronegativity than the adjacent atom draws electrons, resulting in asymmetric electron density. Consequently, this creates a positively charged site on the atom with lower electronegativity. This could induce new chemical and electrical properties such as favourable adsorption of oxygen, enhanced mass transfer capability, and pseudocapacitance, resulting in significantly improved electrochemical performance of supercapacitors and fuel cells.^5,11,31–34

Another strategy to enhance the electrochemical performance of carbon materials for energy storage applications involves doping electrode materials with transition metal (TM) oxides. The availability of various oxidation states of transition metals facilitates the use of four-electron pathways to enhance oxygen reduction reactions (ORR).^35,36 Moreover, the pseudocapacitance provided by TM-oxides enhances the electrochemical performance of supercapacitors.^37,38 Ruthenium oxides (RuO₂) show promise due to their high specific capacitance (720 F g⁻¹), good chemical stability, and high electrical conductivity. However, their cost, toxicity and the rarity of ruthenium limit its use in commercial applications.³⁹ Various cost-effective alternatives to RuO₂ have been developed, including such as Co₃O₄,⁴⁰ MnO₂,⁴¹ NiO⁴² and Fe₃O₄.⁴³ Among such alternatives, manganese oxide (Mn₃O₄) shows promise due to its low toxicity, outstanding electrochemical properties, low cost, and abundance. Unfortunately, it suffers from flaws such as poor conductivity and a low diffusion coefficient,⁴⁴ which originate from the innate structural properties of Mn₃O₄. A promising approach to address the issues with Mn₃O₄ is to synthesize composite electrodes with Mn₃O₄ nanoparticles anchored to highly conductive porous matrix materials that provide more reaction sites between active materials and electrolyte ions. To the best of our knowledge, a hybrid material composed of Mn₃O₄ and boron-doped porous carbon has yet to be investigated as electrode materials for electrochemical systems.

In this work, we have successfully incorporated a certain amount of Mn₃O₄ into boron-doped porous carbon via a two-step process. Boron-doped porous carbons were first synthesized from sodium borohydride via synchronous carbonization of gaseous CO₂. Next, the as-prepared carbon materials were impregnated with Mn(NO₃)₂·4H₂O that was further decomposed to form Mn₃O₄ at an elevated temperature. The electrochemical behaviours arising from boron-doped porous carbon incorporated with Mn₃O₄ promise to enhance the performance of electrocatalysts for oxygen reduction reactions and supercapacitors for energy storage.

Experimental

Materials

Argon (Ar, >99.9%) and carbon dioxide (CO₂, >99.8%) were supplied by Deokyang Co., Ltd. Sodium borohydride (NaBH₄, >99%), manganese nitrate tetrahydrate (Mn(NO₃)₂·4H₂O, ≥97.0%), Nafion® solution (5 wt% in alcohols and 15–20% of water), potassium hydroxide (KOH, 90%), hydrochloric acid (HCl, 37 wt% in water), platinum on graphitized carbon (Pt/C, 20 wt%) and polytetrafluoroethylene (PTFE, 40 wt% in water) were obtained from Sigma-Aldrich. Methanol (CH₃OH, >99.6%) and sodium hydroxide (NaOH, >95%) were purchased from Junsei Chemical Co. Ltd. Ethanol (C₂H₅OH, >99.5%) was acquired from Samchun Chemicals. All chemicals were used without further purification.

Preparation of boron-doped porous carbon (BPC) from CO₂

An alumina crucible boat containing approximately 5 g NaBH₄ was placed into an alumina tube mounted in a furnace (GSL1100X, MTI Co.). The tube was heated to 500 °C (heating rate of 5 °C min⁻¹) for 2 h under CO₂ (flow rate of 76 mL min⁻¹), then cooled down to room temperature. In order to remove impurities and any unreacted reactants, the product was subjected to washing and filtering. The product was washed with 5 M HCl at 90 °C for 1 h, followed by filtering. The filtered product was washed again with 90 mL DI water at 90 °C for 1 h, and then was filtered again. This sequential process of washing and filtering was repeated at 25 °C. Each filtering and washing process was repeated two times. Finally, the product was stirred with ethanol for 30 min and then filtered. After its purification, the product was dried in an oven at 120 °C for 12 h. Generally, an average yield of the carbon from CO₂ is about 9.21% based on the weight of NaBH₄ from four or five repeated experiments.

Synthesis of manganese oxide–BPC composites

To impregnate BPC with manganese oxide, approximately 0.2 g BPC sample was mixed with a manganese precursor (manganese nitrate tetrahydrate) and dissolved in DI water in a vial. The impregnated samples were named BPCMOx, where x is the weight percent of the precursor (10, 20, 30, 40 or 50) based on the weight of BPC. In order to ensure thorough mixing, ultra-sonication was applied to the mixture for 30 min, and then the samples were dried in an oven at 120 °C for 12 h. The dried mixture was placed in the furnace and heated to 850 °C under Ar (flow rate of 76 mL min⁻¹). The furnace temperature was controlled by the following steps: it was raised to 100 °C for the first 30 min and maintained for 1 h. Afterwards, the tube was heated again to 850 °C (heating rate of 5 °C min⁻¹), its temperature maintained for 2 h, and then finally cooled to room temperature. To remove impurities, the synthesized products were purified by washing and filtering with DI water and ethanol. The products were then dried in an oven at 120 °C for 12 h. The resultant products were Mn₃O₄-impregnated BPC of increasing wt% of precursor (BPCMO10, 20, 30, 40 and 50).

Physical characterization

Scanning electron microscopy (SEM) images were obtained with a Magellan 400 UHR-SEM at 1–2 kV. Transmission electron microscopy (TEM) images were obtained on a Tecnai G2 F30 operating at 300 kV. Powder X-ray diffraction (XRD) was conducted using a D/MAX-2500 with Cu Kα (λ = 0.15418 nm), operating at 40 kV and 200 mA for a 2θ range of 10 to 90° at a scan rate of 1° min⁻¹, and step size of 0.02°. The nitrogen adsorption/desorption isotherms were obtained at 77 K on a Micromeritics 3-FLEX after the samples were degassed under vacuum at 200 °C for 6 h. The specific surface area was calculated according to the Brunauer–Emmett–Teller (BET) method and the pore size distribution was determined from a non-local density functional theory (NLDFT) method. Raman spectroscopic analyses were examined using a high-resolution dispersive Raman microscope (ARAMIS) by Horiba Jobin Yvon, with an argon ion CW laser (514.5 nm) as the excitation source. X-ray photoelectron spectroscopy (XPS) data obtained using a Sigma probe (Thermo VG Scientific) was fitted with Avantage software (Thermo VG package) and the binding energy was referenced to the C 1s at 284.5 eV.

Oxygen reduction reaction (ORR) measurements

The ORR electrochemical measurements consisted of cyclic voltammetry (CV), rotating ring-disk electrode (RRDE) and rotating disk electrode (RDE) measurements. A potentiometer (Biologic) was used to perform these analyses. The measurements were performed in 1 M NaOH electrolyte saturated by oxygen. It was evaluated using a three-electrode system at room temperature. The electrodes used were Ag/AgCl (reference electrode), platinum wire (counter electrode), glassy carbon disk (3 mm diameter, working electrode for CV and RDE measurement) and Pt ring/GC disk (working electrode for RRDE measurement). For the measurements, inks containing 5 mg of the sample dispersed in methanol (250 μL), water (750 μL) and Nafion® solution (50 μL) were prepared by ultra-sonication (for 35 min). After 3.5 μL of the solution was deposited on the polished working electrode, the treated electrode was dried for 12 h in air.

Supercapacitor electrochemical measurements

Electrochemical measurements consisted of cyclic voltammetry (CV), galvanostatic charge/discharge and electrochemical impedance spectroscopy (EIS). They were performed in 6 M KOH aqueous solution electrolyte at room temperature and measured using a three-electrode system with saturated calomel electrode (SCE, reference electrode) and platinum wire (counter electrode). The working electrode was prepared by mixing BPCMO and PTFE (ca. 2.5 wt% in water, as a binder) with a mass ratio of 9 : 1 in DI water. Approximately 4 mg of slurry was loaded onto 1 × 1 cm nickel foam (as a current collector), which was then pressed and dried in an oven at 120 °C for 12 h. A potentiometer (Biologic) was used for these analyses. To control the potential at the working electrode, a CHI 6002D instrument was employed.

Results and discussion

Morphological characteristics

The structure and morphology of the resulting products were investigated using scanning electron microscopy (SEM) (Fig. 1). Fig. 1a is BPC, the reference sample, and Fig. 1b is BPCMO40, the manganese oxide-impregnated BPC. The surface of BPCMO40 seems to be rougher than that of BPC. Fig. 1c is a back-scattered electron (BSE) image of Fig. 1b, showing the presence of manganese oxide within the BPC framework (brighter portions indicate manganese oxide nanoparticles). These images indicate that the manganese oxide nanoparticles are uniformly dispersed in BPC, which can also be confirmed in the transmission electron microscopy (TEM) images shown in Fig. 2. The TEM images of BPC and BPCMO40 in Fig. 2 also show that the manganese oxide (black dots) is well placed within the BPC scaffold. The inset image in Fig. 2b shows at higher magnification that the manganese oxide is embedded in the BPC. This suggests mutual interaction between carbon and manganese oxide that might affect the electrochemical performance of BPCMO40.

Fig. 1 SEM images of (a) BPC, (b) BPCMO40 and (c) BSE image of (b).

Fig. 2 TEM images of (a) BPC (scale bar: 50 nm), (b) BPCMO40 with the inset showing BPCMO40 at higher magnification (scale bar: 100 nm).

To identify the specific type of manganese oxide and obtain more detailed structural information, X-ray diffraction (XRD) measurements were performed for BPC and BPCMO40 (Fig. 3 and S1 in ESI†). Of all the possible forms of manganese oxide, the BPCMO40 contains only Mn₃O₄, as evidenced by the distinct peaks indexed to a tetragonal spinel (JCPDS no. 24-0734) in the profile.⁴⁵ In view of the carbon structure, the existence of amorphous carbon is evidenced by a broad peak at 2-theta values of ca. 25 for BPC, but the peak becomes slightly sharpened for BPCMO40. Thus, the BPCMO40 has more graphitized carbon structure after the impregnation and subsequent heat-treatment step. The result of Raman spectroscopy, in which I_D/I_G increases in graphitization of the amorphous carbon case, also supports this (Fig. S2, refer to ESI†).^25,46 The value of I_D/I_G is 0.92 (BPC), 0.94 (BPCMO30), 0.94 (BPCMO40) and 0.97 (BPCMO50). It seems that thermal energy mainly contributes to the more graphitized carbon structure by comparing with control group (named as BPC850) which is heat-treated at 850 °C without the manganese precursor. Based on our previous work, the I_D/I_G value of BPC850 is 1.10.²⁵ Interestingly, the formation of Mn₃O₄ tends to involve restraining the graphitization of the carbon structure because BPCMO has a lower value than that of BPC850.

Fig. 3 XRD patterns of BPC and BPCMO40. (Filled squares and open squares indicate peaks of Mn₃O₄ and amorphous carbon, respectively.)

To investigate the porosity of BPCMOx, N₂ adsorption/desorption isotherm measurements were performed. In Fig. 4, all the isotherms exhibit type IV isotherm, with a mixed H1 and H3 hysteresis loop in the range of the relative pressure (0.25–1.0). This comes from capillary condensation phenomena occurring in mesopores.⁴⁷ The wide relative pressure range indicates that the BPCMO materials have mesopores of various sizes. In addition, the condensation effect observed at higher pressure reveals the existence of macropores.⁴⁸ Fig. 5 shows that the pore size distribution (PSD) of BPC and BPCMOs is in the range 0.5–99 nm, which implies the presence of micro, meso, and macropores. This range would allow internal surfaces to be wet with the electrolyte and to provide active sites for electrochemical reactions,⁴⁹ facilitating the transportation of ions and electrons.⁵⁰ The decreasing order of nitrogen sorption amount is BPCMO30, BPCMO50, BPCMO40 and BPC, which is in accordance with the tendency of specific surface area (SSA) (Table 1). When BPC is impregnated with manganese oxide, the BET surface area (S_BET) increases from 515 (BPC) to 615 (BPCMO30) m² g⁻¹ due to the thermal oxidation process of the manganese precursor.⁵¹ The PSD in Fig. 5 shows identical profiles because all of the materials are variants of BPC. However, this is not the case for pore sizes of 0.59 and 0.77 nm. In fact, the impregnated Mn₃O₄ has greater pore volumes at these values, leading to higher SSA values than the SSA of BPC, which increases the surface area for ion adsorption.⁵⁰ To ensure that this was not due to thermal effects, BPC was annealed at 850 °C for 2 h, the same synthesis conditions as for BPCMOx. The S_BET of BPC was greater than the S_BET of BPC850 (Table 1), demonstrating that the increase of SSA in the BPCMOx originates from the impregnation of Mn₃O₄.

Fig. 4 Nitrogen adsorption/desorption isotherms of BPC and BPCMOs.

Fig. 5 PSD of BPC and BPCMOs based on NLDFT model.

Table 1 Element distribution and porosity parameters obtained from the XPS analysis and nitrogen adsorption/desorption isotherm for the BPC and BPCMOs

	Initial mass ratio of Mn precursor	C (at%)	O (at%)	B (at%)	Mn (at%)	SBET (m^2 g^−1)	Vtotal (cm^3 g^−1)
a XPS data from Byeon et al.^25
BPC850^a	—	82.77	11.45	5.78	—	137	0.63
BPC	—	64.03	16.09	19.88	—	515	0.85
BPCMO30	0.3	88.81	9.23	0.48	1.47	615	0.87
BPCMO40	0.4	74.36	17.1	1.84	6.7	563	0.80
BPCMO50	0.5	81.6	12.28	1.99	4.13	598	0.80

---

In order to understand the composition and chemical bonding states in BPC and BPCMOx, X-ray photoelectron spectroscopy (XPS) was performed. Table 1 shows the elemental distribution from the XPS survey data, which supports the presence of B, C, O and Mn atoms. According to the result of B 1s, as the ratio of Mn precursor increases, the atomic percentage of boron increases, indicating the existence of correlation between boron and manganese. Fig. 6a displays the spectrum for Mn 2p in BPCMO40, where distinct peaks at Mn 2p_3/2 (641.7 eV) and Mn 2p_1/2 (653.3 eV) are clearly observable. Here, a notable difference of 11.6 eV can be observed, proving the presence of Mn₃O₄. In addition, a slight shift (left) toward carbon is observed. This implies some interaction between Mn₃O₄ and carbon within BPCMO40, as evidenced in the manganese–carbon composites.³⁴ The deconvolution of the B 1s spectrum of BPC and BPCMOs (Fig. 6b) indicates the presence of oxygen and boron containing functional groups such as B₄C (187.5 eV), BC₃ (188.9 eV), BC₂O (191 eV) and BCO₂ (192.3 eV).^24,27 Impregnation of the BPC with Mn₃O₄ affects the bonding types of boron: as more Mn₃O₄ is added, B₄C and BC₃ tend to decrease while the portion of BCO₂ or BC₂O increases (Fig. 6c and Table 2), implying increased oxidation of boron.²⁵

Fig. 6 XPS (a) Mn 2p of BPCMO40, (b) B 1s of BPC and BPCMOs and (c) stack column of (b).

Table 2 List of the oxygen and boron containing functional groups based on the deconvolution of the B 1s spectrum of BPC and BPCMOs

	B4C (at%)	BC3 (at%)	BC2O (at%)	BCO2 (at%)
BPC	83.24	15.00	1.76	—
BPCMO30	45.60	18.62	19.79	15.98
BPCMO40	15.83	20.39	38.30	25.48
BPCMO50	8.67	9.18	78.41	3.74

---

Electrochemical oxygen reduction reaction (ORR) activity

The electrochemical activity of BPCMOs for ORR was evaluated by cyclic voltammetry (CV) in 1 M NaOH solution. This analysis showed enhanced ORR electrocatalytic activity from the synergistic effects of B and Mn₃O₄. As demonstrated in Fig. 7a, a cathodic reduction peak was detected at −0.34 V (vs. Ag/AgCl) in the O₂-saturated electrolyte for BPC. This is equivalent to 0.66 V relative to the standard reversible hydrogen electrode (RHE, refer to the ESI† for the RHE conversion equation). The peak potential (E_peak) of BPCMOs is −0.21 V (0.79 V vs. RHE), independent of the amount of Mn₃O₄ impregnation. However, there is a shift in the positive direction upon Mn₃O₄ impregnation, closing the gap between the potential of Pt/C (20 wt%) (0.86 V vs. RHE). Comparing this to Pt/C (20 wt%), the E_peak of the BPCMOs is more negative but the current densities (I_peak) were much higher than that of Pt/C (I_peak (BPCMO40) = −3.69 mA cm⁻², I_peak (Pt/C) = −1.81 mA cm⁻²). In fact, the performance indicates higher catalytic activity than that of boron-doped graphene.⁵² Furthermore, BPCMO40 has a more stable current density than Pt/C in the CV range of 0 to −1 V, revealing the improved electrocatalytic activity of BPCMO40.

Fig. 7 CV of (a) BPC, BPCMOx and Pt/C (20 wt%); (b) BPC, BPCMO40 and Pt/C (20 wt%) in Ar (dot line) and O₂ (solid line) saturated NaOH at a voltage scan rate of 50 mV s⁻¹.

Fig. 7b shows the CV plots of BPC, BPCMO40 and Pt/C (20 wt%) in Ar and O₂-saturated 1 M NaOH solution. The current generated from ORR, the difference in current between Ar and O₂ gas, is higher than that of Pt/C. The rise in current density implies an increment in catalyzed ORR, which is attributed to the increased number of active sites exposed to oxygen molecules.³⁶ However, in the presence of Ar a reduction peak was not observed, but a secondary peak at approximately −0.4 V (vs. Ag/AgCl) appeared. This peak was also observed in the presence of O₂, and we attribute this to a reaction between metal oxide and electrolyte.⁵³ Mn₃O₄ causes a redox reaction, as do the other manganese oxides. The charge storage reaction of Mn₃O₄ is very similar to that of MnO₂.^54,55 Thus, Na⁺ in the electrolyte may lead to the following reduction reaction of Mn₃O₄:

MnO_1.33 + δNa⁺ + δe⁻ ↔ MnO_1.33Na_δ (1)

Fig. 8 is a voltammogram of an RDE at a speed of 2500 rpm for BPC, BPCMOs and Pt/C (20 wt%). For less negative potentials, BPCMOs show abrupt increases in current compared to Pt/C (20 wt%). The half-wave potential (E_1/2) for ORR (the ORR initiation potential) is 0.857 V (vs. RHE) for Pt/C, 0.803 V (vs. RHE) for BPCMO30, 0.805 V (vs. RHE) for BPCMO40 and 0.826 (vs. RHE) for BPCMO50. The co-utilization of B and Mn₃O₄ results in a positive shift from 0.657 in BPC to 0.826 in BPCMO50. The current density at −0.3 V (0.7 V vs. RHE, a common half-cell cathodic potential of a fuel cell (I_{−0.3 V})) of BPCMO40 is approximately −4.633 mA cm⁻², a higher value than that of Pt/C (20 wt%) (−4.16 mA cm⁻²). The increase in available active sites contributed to the increased current density.

Fig. 8 LSV of BPC, BPCMOs and Pt/C (20 wt%) at 2500 rpm and at a scan rate of 50 mV s⁻¹.

The electron transfer number (n) and the amount of hydrogen peroxide (H₂O₂) produced by each product were gained from RRDE measurements in O₂-saturated 1 M NaOH electrolyte (see the ESI† for details of the calculations). As shown in Fig. 9, BPCMO40 exhibits the highest ORR activity among the BPCMOs in the entire range of −0.3 V to −1.0 V. The range of n for BPCMO40 is 3.5–3.8, which is quite close to that of Pt/C in the range of potential between −0.3 V and −1.0 V, indicating that it follows a four-electron transfer pathway. Unlike BPCMOx, the electron transfer number is 1.8–2.4 for bare BPC in the range of potential between −0.3 V and −1.0 V, indicating a two-electron transfer pathway, an undesired mechanism for ORR due to its slow reaction and peroxide generation. It is clear that the addition of Mn₃O₄ to the carbon plays a key role in enhancing the electrocatalytic activities. As shown in Fig. 9b, there exists an inverse relationship between n and H₂O₂ yield.

Fig. 9 Electron transfer number (n) and the amount of peroxide (HO₂⁻) produced, calculated from RRDE data at 2500 rpm and at a scan rate of 50 mV s⁻¹.

The values for CV, RDE and RRDE are summarized in Table 3. When BPC was impregnated with Mn₃O₄, the E_peak, I_peak, E_1/2, I_{−0.3 V} and n_{−0.3 V} values all increased. The ratio of oxidized boron increases, carbon and manganese begin to interact, and SSA increases. When combined, these results indicate the improved ORR performance is explained by an increase in the number of active sites. In particular, the results obtained for BPCMO40 were comparable to Pt/C. XPS results (Table 2) for BPCMO40 show that the most oxidized form of boron, BCO₂, is present in significant amounts, accounting for the high ORR activity.²⁵ Thus, the synergy between boron-doping and manganese oxide impregnation leads to more active sites from oxidized boron, ultimately showing highly enhanced electrocatalytic performance.

Table 3 Comparison of the ORR catalytic performances

Samples	Epeak (V)	Ipeak (mA cm^−2)	E1/2 (V)	I−0.3 v (mA cm^−2)	n−0.3 v
BPC	0.66	−0.39	0.657	−0.261	1.79
BPCMO30	0.79	−2.69	0.803	−3.863	3.01
BPCMO40	0.79	−3.69	0.805	−4.633	3.48
BPCMO50	0.79	−2.12	0.826	−4.067	3.15
Pt/C (20 wt%)	0.86	−1.81	0.857	−4.16	3.96

---

Fig. 10a is the result of stability tests for BPCMO40 and Pt/C (20 wt%). Using the chronoamperometry technique, the test was performed at −0.7 V in O₂-saturated 1 M NaOH. The current density of BPCMO40 decreased by 9.5% for 24 000 s, whereas that of Pt/C decreased by about 20%. We also tested this material in the presence of methanol to check its tolerance to methanol crossover effects (Fig. 10b). LSV for Pt/C (20 wt%) shows huge oxidation current density at about −0.3 V on the LSV curve. However, BPCMO40 does not show oxidation current density, suggesting its tolerance to the alcohol fuel.

Fig. 10 (a) Current–time chronoamperometric responses of BPCMO40 and Pt/C (20 wt%) in O₂ saturated 1 M NaOH at 1200 rpm; (b) LSV of BPCMO40 and Pt/C (20 wt%) in O₂ saturated 1 M NaOH (80 mL) with 3 M methanol (5 mL) at 1600 rpm and a scan rate of 50 mV s⁻¹.

Supercapacitor electrochemical analyses

To investigate the electrochemical behaviour for supercapacitor applications, cyclic voltammetry (CV) was performed. Fig. 11 shows the CV curves and CV was performed at a scan rate of 10 mV s⁻¹ between 0 and −1.0 V (vs. SCE) using BPC and BPCMOs. The CV curves show roughly rectangular images, which indicates that the BPCMO electrode behaves as an ideal capacitor in the range of potential 0 to −1.0 V. The area under the CV curve of BPCMO50 is the largest, indicating that its specific capacitance (SC) value is the highest. This is because SC is proportional to the area under the current density–voltage curve at the same scan rate and potential window.⁶ There are redox peaks in the range of potential 0 to −1.0 V, and the pseudocapacitance possibly originates from the redox reaction between manganese oxide and electrolyte as mentioned before in the ORR section. K⁺ in the supercapacitor electrolyte acts like Na⁺ in the ORR experiment.

Fig. 11 CV curves of BPC and BPCMOx with the range of potential −1.0 to 0.0 V, at a scan rate of 10 mV s⁻¹.

Galvanostatic charge/discharge (GCD) measurements were carried out to acquire more information on the capacitive performance of the BPCMOs. Fig. 12a shows the GCD curves of BPC and BPCMOs at a current density of 1 A g⁻¹ in 6 M KOH electrolyte solution in the potential range of −1.0 to 0.0 V (vs. SCE). The increase in the charging and discharging time indicates that the capacitance of the BPCMO composite is greater than that of BPC, which agrees with the CV results. As shown in Fig. 12b, SC values were 59, 107, 118 and 150 F g⁻¹ at 1 A g⁻¹ for BPC, BPCMO30, 40 and 50, respectively (refer to the ESI† for the calculation of SC). The BPCMO50 retained the highest capacitance due to its pseudocapacitance. The SC for BPCMO50 was 150, 147, 142, 135, 133 and 136 F g⁻¹ at current densities of 1, 2, 3, 5, 7 and 10 A g⁻¹, respectively. BPCMO50 maintained almost constant SC at various current densities, indicating its high stability. In addition, the BPCMO series showed outstanding normalized capacitance (specific capacitance divided by the surface area): 17.4 (BPCMO30), 21.0 (BPCMO40) and 25.1 μF cm⁻² (BPCMO50), which is better than that of the graphene-based porous carbon (∼9 μF cm⁻²).⁵⁶

Fig. 12 (a) GCD curves at 1 A g⁻¹ in 6 M KOH and (b) SC of BPC and BPCMOs at different current density.

For better understanding of the superior capacitive performance of BPCMOs, electrochemical impedance spectroscopy (EIS) analyses were conducted. The Nyquist plot was presented in Fig. 13. According to the analysis, the Nyquist plot is composed of a semi-circular part and a vertical part. In the high-frequency region, the intercept at the beginning of the semicircle is related to the electrical resistance of the electrolyte (R_s). The R_s is 0.275, 0.336 and 0.345 Ω for BPCMO40, 50 and 30, respectively. In addition, the semicircle radius corresponds to the faradaic charge transfer resistance (R_ct). As shown in Fig. 13, the R_ct of BPCMO40, 50 and 30 was 0.257, 0.259 and 0.217 Ω, respectively. Both R_s and R_ct values of BPCMOx were very low, compared to the results from prior work focused on composites of graphene–Mn₃O₄ (R_s = 2.02 Ω, R_ct = 1.06 Ω), carbon nanotube–Mn₃O₄ (R_s = 1.25 Ω, R_ct = 1.1 Ω) and CNF–Mn₃O₄ (R_s > 4 Ω, R_ct > 2 Ω) for supercapacitor electrodes.^45,57,58 In the low-frequency range, the linear part is related to the ion diffusion and transport in the electrolyte. For BPCMOx, the slope of the plot was steeper than for BPC, demonstrating that BPCMOx have faster diffusion of electrolyte ions within the pores of electrodes during redox reactions, which leads to better capacitive behaviour. Furthermore, the long-term cycling performance of the BPCMO50 was performed by the GCD at a current density of 2 A g⁻¹ (Fig. 14). During 3500 cycles, only 6.1% capacitance loss was observed, indicating its viability for practical applications. The unchanged SEM image (Fig. S11 in ESI†) of the spent sample after the cycling test also supports the high stability of the composite.

Fig. 13 Nyquist plots of BPC, BPCMOs using a sinusoidal signal of 10 mV over the frequency range of 100 kHz to 0.05 Hz.

Fig. 14 Cycle stability of BPCMO50 at the current density of 2 A g⁻¹.

Conclusions

Boron-doped porous carbon impregnated with Mn₃O₄ (BPCMO) has been synthesized with a facile two-step method of CO₂ conversion at 500 °C and subsequent annealing at 850 °C. The incorporation of boron and Mn₃O₄ into the carbon support provides more active sites through the interactions between oxidized boron/manganese and carbon. Moreover, the specific surface area increases, and the total resistance of the nanocomposite (BPCMO) decreases, after incorporation of the metal oxide. As a result, BPCMO40 shows significantly improved ORR activity and features excellent tolerance to methanol. BPCMO50 exhibits enhanced electrochemical performance, such as high capacitance and good stability, for use in supercapacitors for energy storage. Therefore, the BPCMOx series could provide viable candidates to serve as alternatives to Pt electrodes for ORR, and as electrode materials for supercapacitors.

Acknowledgements

The authors are grateful for financial support from the UK-Korea Joint Research Program through NRF grants (NRF-2015M2A7A1000219) and the Korea CCS R & D Center funded by the Ministry of Science, ICT and Future Planning (NRF-2014M1A8A1049297).

Notes and references

1. D. Y. Chung, K. J. Lee, S.-H. Yu, M. Kim, S. Y. Lee, O.-H. Kim, H.-J. Park and Y.-E. Sung, Adv. Energy Mater., 2015, 5, 1401309.
2. V. V. Galvita, H. Poelman and G. B. Marin, J. Power Sources, 2015, 286, 362–370.
3. A. Allagui, A. H. Alami, E. A. Baranova and R. Wüthrich, J. Power Sources, 2014, 262, 178–182.
4. B. S. Yin, S. W. Zhang, H. Jiang, F. Y. Qu and X. Wu, J. Mater. Chem. A, 2015, 3, 5722–5729.
5. Z.-J. Jiang, Z. Jiang and W. Chen, J. Power Sources, 2014, 251, 55–65.
6. S. M. Chen, R. Ramachandran, V. Mani and R. Saraswathi, Int. J. Electrochem. Sci., 2014, 9, 4072–4085.
7. L. Hao, X. Li and L. Zhi, Adv. Mater., 2013, 25, 3899–3904.
8. J. Cho, S. Jeong and Y. Kim, Prog. Energy Combust. Sci., 2015, 48, 84–101.
9. Z. Jiang, B. Pei and A. Manthiram, J. Mater. Chem. A, 2013, 1, 7775–7781.
10. C. Vix-Guterl, E. Frackowiak, K. Jurewicz, M. Friebe, J. Parmentier and F. Beguin, Carbon, 2005, 43, 1293–1302.
11. S. J. Yuan and X. H. Dai, RSC Adv., 2015, 5, 45827–45835.
12. A. L. Dicks, J. Power Sources, 2006, 156, 128–141.
13. J. Jiang, J. H. Zhu, W. Ai, Z. X. Fan, X. N. Shen, C. J. Zou, J. P. Liu, H. Zhang and T. Yu, Energy Environ. Sci., 2014, 7, 2670–2679.
14. X. Zhang, Z. H. Wei, Q. J. Guo and H. J. Tian, J. Power Sources, 2013, 231, 190–196.
15. G. F. Ma, D. Y. Guo, K. J. Sun, H. Peng, Q. Yang, X. Z. Zhou, X. L. Zhao and Z. Q. Lei, RSC Adv., 2015, 5, 64704–64710.
16. L. Wang, L. C. Zhang, J. X. Cheng, C. X. Ding and C. H. Chen, Electrochim. Acta, 2013, 102, 306–311.
17. M. P. Yu, R. Li, Y. Tong, Y. R. Li, C. Li, J. D. Hong and G. Q. Shi, J. Mater. Chem. A, 2015, 3, 9609–9615.
18. Y. Xia, Z. Xiao, X. Dou, H. Huang, X. H. Lu, R. J. Yan, Y. P. Gan, W. J. Zhu, J. P. Tu, W. K. Zhang and X. Y. Tao, ACS Nano, 2013, 7, 7083–7092.
19. J. Zhang, Y. Zhao, X. Guan, R. E. Stark, D. L. Akins and J. W. Lee, J. Phys. Chem. C, 2012, 116, 2639–2644.
20. J. Zhang and J. W. Lee, Carbon, 2013, 53, 216–221.
21. R. Xiong, X. Li, A. Byeon and J. W. Lee, RSC Adv., 2013, 3, 25752–25757.
22. J. Zhang, Y. Zhao, D. L. Akins and J. W. Lee, J. Phys. Chem. C, 2010, 114, 19529–19534.
23. J. S. Zhang and J. W. Lee, ACS Sustainable Chem. Eng., 2014, 2, 735–740.
24. J. Zhang, A. Byeon and J. W. Lee, J. Mater. Chem. A, 2013, 1, 8665–8671.
25. A. Byeon, J. Park, S. Baik, Y. Jung and J. W. Lee, J. Mater. Chem. A, 2015, 3, 5843–5849.
26. A. Byeon and J. W. Lee, J. Phys. Chem. C, 2013, 117, 24167–24173.
27. S. Baik and J. W. Lee, RSC Adv., 2015, 5, 24661–24669.
28. X. Zhao, Q. Zhang, B. Zhang, C.-M. Chen, J. Xu, A. Wang, D. S. Su and T. Zhang, RSC Adv., 2013, 3, 3578–3584.
29. H. Z. Chi, Y. Li, Y. Xin and H. Qin, Chem. Commun., 2014, 50, 13349–13352.
30. J. Song, Y. Zhang and Y. Liu, RSC Adv., 2015, 5, 20734–20740.
31. U. B. Nasini, V. G. Bairi, S. K. Ramasahayam, S. E. Bourdo, T. Viswanathan and A. U. Shaikh, J. Power Sources, 2014, 250, 257–265.
32. K. T. Cho, S. B. Lee and J. W. Lee, J. Phys. Chem. C, 2014, 118, 9357–9367.
33. Z. Yang, Z. Yao, G. Li, G. Fang, H. Nie, Z. Liu, X. Zhou, X. A. Chen and S. Huang, ACS Nano, 2012, 6, 205–211.
34. J. Duan, Y. Zheng, S. Chen, Y. Tang, M. Jaroniec and S. Qiao, Chem. Commun., 2013, 49, 7705–7707.
35. Y. Gorlin, C.-J. Chung, D. Nordlund, B. M. Clemens and T. F. Jaramillo, ACS Catal., 2012, 2, 2687–2694.
36. G. M. Kim, S. Baik and J. W. Lee, RSC Adv., 2015, 5, 87971–87980.
37. B. G. S. Raj, R. N. R. Ramprasad, A. M. Asiri, J. J. Wu and S. Anandan, Electrochim. Acta, 2015, 156, 127–137.
38. H. Wang, Y. Q. Fu, X. Y. Wang, J. Gao, Y. W. Zhang and Q. L. Zhao, J. Alloys Compd., 2015, 639, 352–358.
39. J. P. Zheng, P. J. Cygan and T. R. Jow, J. Electrochem. Soc., 1995, 142, 2699–2703.
40. H. T. Cui, M. M. Wang, W. Z. Ren and Y. A. Zhao, Funct. Mater. Lett., 2014, 7, 1450002.
41. M. Lu, S. Kharkwal, H. Y. Ng and S. F. Y. Li, Biosens. Bioelectron., 2011, 26, 4728–4732.
42. A. Trunov, Electrochim. Acta, 2013, 105, 506–513.
43. X. A. Du, C. Y. Wang, M. M. Chen and Y. Jiao, J. Inorg. Mater., 2008, 23, 1193–1198.
44. B. H. Shambharkar, S. S. Umare and R. C. Rathod, Trans. Indian Inst. Met., 2014, 67, 827–834.
45. G. Z. Jin, X. X. Xiao, S. Li, K. M. Zhao, Y. Z. Wu, D. Sun and F. Wang, Electrochim. Acta, 2015, 178, 689–698.
46. A. C. Ferrari and J. Robertson, Phys. Rev. B: Condens. Matter Mater. Phys., 2000, 61, 14095–14107.
47. H. T. Zhang, X. Zhang and Y. W. Ma, Electrochim. Acta, 2015, 184, 347.
48. D. D. Zhu, L. J. Li, J. J. Cai, M. Jiang, J. B. Qi and X. B. Zhao, Carbon, 2014, 79, 544–553.
49. A. Pramanik, S. Maiti and S. Mahanty, Dalton Trans., 2015, 44, 14604–14612.
50. D. S. Achilleos and T. A. Hatton, J. Colloid Interface Sci., 2015, 447, 282–301.
51. C. Jiang, J.-M. Séquaris, A. Wacha, A. Bóta, H. Vereecken and E. Klumpp, Geoderma, 2014, 235–236, 260–270.
52. Y. Zhou, C. H. Yen, S. Fu, G. Yang, C. Zhu, D. Du, P. C. Wo, X. Cheng, J. Yang, C. M. Wai and Y. Lin, Green Chem., 2015, 17, 3552–3560.
53. C. F. Liu, H. Q. Song, C. K. Zhang, Y. G. Liu, C. P. Zhang, X. H. Nan and G. Z. Cao, Nano Res., 2015, 8, 3372–3383.
54. W. Wei, X. Cui, W. Chen and D. G. Ivey, Chem. Soc. Rev., 2011, 40, 1697–1721.
55. M. Q. Wu, G. A. Snook, G. Z. Chen and D. J. Fray, Electrochem. Commun., 2004, 6, 499–504.
56. P. Yadav, A. Banerjee, S. Unni, J. Jog, S. Kurungot and S. Ogale, ChemSusChem, 2012, 5, 2159–2164.
57. K. B. Wang, X. B. Shi, A. M. Lu, X. Y. Ma, Z. Y. Zhang, Y. N. Lu and H. J. Wang, Dalton Trans., 2015, 44, 151–157.
58. X. Zhao, Y. Du, Y. Li and Q. Zhang, Ceram. Int., 2015, 41, 7402–7410.

---

Footnote

† Electronic supplementary information (ESI) available: Raman, XRD and electrochemical analysis data of BPCMOs. See DOI: 10.1039/c6ra10061a

---